INDEX

Explanations

phrases related to hiding or concealing information or identity

oai_token-act-pair · gpt-3.5-turbo

references to concealment or hidden identities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inelli

-0.77

usa

-0.69

HAM

-0.69

starter

-0.68

ularity

-0.67

lishes

-0.67

review

-0.67

rano

-0.66

 roundup

-0.66

grounds

-0.66

POSITIVE LOGITS

 secrets

1.31

 wrongdoing

1.12

 whereabouts

1.06

 truths

1.04

 identities

0.98

 secret

0.98

 hidden

0.97

 identity

0.92

hid

0.92

 treacher

0.91

Activations Density 0.386%

phrases related to hiding or concealing information or identity

references to concealment or hidden identities

No Comments

No Known Activations